EVALUATION OF CaCl 2–SILICA GEL SORBENT FOR WATER …mbahrami/pdf/2015/Evaluation of CaCl2-silica gel sorb… · In the case of salt in silica gel composites, ... 200oC for six

Heat Pipe Science and Technology: An International Journal 6(3−4) 229−239 (2015)

EVALUATION OF CaCl2–SILICA GELSORBENT FOR WATER SORPTIONCOOLING SYSTEMS

Claire McCague, Khorshid Fayazmanesh, Cecilia Berlanga,& Majid Bahrami ∗

Laboratory for Alternative Energy Conversion (LAEC), School of MechatronicSystems Engineering, Simon Fraser University, 250-13450 102nd Avenue,Surrey, BC, Canada, V3T 0A3

*Address all correspondence to: Majid Bahrami; E-mail: [email protected]

Hygroscopic salts supported by a mesoporous matrix for improved mass transport are promisingsorbents for water-based sorption cycles that operate at low pressure. In this study, loose graincomposites of CaCl2 supported by mesoporous silica gels with four distinct pore size distributionswere prepared and compared with AQSOA FAM-Z02, a silicoaluminophosphate zeolite desiccant.A salt in silica gel sorbent consolidated with graphite flakes and binder was also analyzed. Thesorbents were evaluated with a volumetric nitrogen physisorption porosimeter and thermo-gravimetric water vapor sorption analyzer. The hygroscopic salt filled 56–60% of the open porevolume of the mesoporous silica gel supports. Water uptake capacity of the CaCl2/silica gel sor-bent was up to 0.33 g per gram of dry sorbent at 12 mbar and 35oC, and sample performancewas consistent through 200 wetting–drying cycles.

KEY WORDS: calcium chloride, sorption cooling, water isotherm

1. INTRODUCTION

The vapor compression systems commonly used for air conditioning and refrigera-tion consume significant amounts of electrical power and employ environmentallyharmful refrigerants. The power used by HVAC systems to provide thermal com-fort in residential and commercial buildings amounts to 10–20% of the total energyconsumed in the developed world (Cullen and Allwood, 2010; Isaac and vanVuuren, 2009; Pérez-Lombard et al., 2008). Efficient, sustainable cooling technolo-gies are presently the focus of significant research and development efforts. In ad-sorption cooling systems, evaporative cooling power is created as a porous sorbentmaterial, which can be regenerated with thermal energy, adsorbs or absorbs a re-frigerant, such as water. Sorption cooling systems that utilize materials with low regeneration temperaturescan be powered by low-grade heat sources, including solar thermal and industrial

2151-7995/15/$35.00©2015 by Begell House, Inc. 229

waste heat (Wang and Oliveira, 2006).The selection of an effective sorbate–sorbentworking pair for a specific adsorption system depends on the intended operating con-ditions, particularly the temperatures and working pressures for adsorption and desorp-tion. Zheng et al. (2014) and Aristov (2013a) reviewed solid desiccants for sorptioncooling systems and compared salt in a host matrix composites with other classesof sorbents, including silica gels, zeolites, aluminophosphates, porous polymericmetal–organic frameworks, and porous carbons. Both reviews conclude that salt inporous matrix sorbents are strong candidate materials with properties that can betuned through selection of the host material and the salt. The focus of this research is the preparation and evaluation of sorbent materialsfor water sorption systems with a target regeneration temperature of 80oC, particu-larly CaCl2–silica gel sorbents. This sorbent type was chosen for its uptake capac-ity and uptake rate under the desired operating conditions. Glaznev et al. (2011)compared SWS-1L, a CaCl2 in SBA-15 silicate composite with microporous sil-ica, and AQSOA FAM-Z02 silicoaluminophosphate by the large-temperature jumpmethod and determined that it had a superior instantaneous specific coolingpower. The sorbent choice is also motivated by the availability of affordable saltsand silica gels, and the ease of synthesis of batches ranging from grams of mate-rial for characterization to kilograms of material for testing in a lab-scale sorptionchiller. Water-based sorption chillers operate at low pressures and sorption dynamicshave proven to be very sensitive to the presence of even a fraction of a millibarof background gas, as was demonstrated by Glaznev et al. (2010) in tests of silicagel, CaCl2 in silica gel, and Z02. Sorbents containing CaCl2 can contribute to thecorrosion of metal heat exchangers, particularly mixed metal heat exchangers sus-ceptible to galvanic corrosion. In our lab-scale sorption chiller tests, we observedperformance loss due gas released by the corrosion of a copper–aluminum sorberbed heat exchanger after a single desorption cycle and the restoration of perform-ance with the evacuation of this background gas between cycles. However, a sorp-tion chiller with CaCl2 in mesoporous silica gel coated on an aluminum alloy heatexchanger was run for 100 cycles by Freni et al. (2007) without any trace of cor-rosion or performance loss. For loose grain sorbent-heat exchanger designs, sorption kinetics studies have re-vealed that for grain sizes below 0.5–0.8 mm the kinetic curves do not change ap-preciably, for example, one layer of 0.8 mm silica gel grains can deliver the samespecific cooling power as four layers of 0.2 mm grains (Sapienza et al., 2014;Aristov, 2013b; N’Tsoukpoe et al., 2014). The sorption rate depends on the trans-port of water vapor in the porous matrix and the dissipation of the heat of adsorp-tion. In the case of salt in silica gel composites, absorption of water vapor by thesalt dispersed in the porous matrix creates a solid hydrate, and then an aqueous so-lution (Ovoshchnikov et al., 2011). The solution formation complicates sorption ki-netics compared to water sorbed by silica gel or zeolite surfaces.

Heat Pipe Science and Technology: An International Journal

230 McCague et al.

The salt confined in the support matrix undergoes water absorption, dissolution,and dilution. The density of the solution depends on the temperature and the massfraction of CaCl2. The enthalpy of adsorption is also concentration dependent.Conde (2004) reviewed the properties of aqueous solutions of CaCl2 and LiCl togenerate and validate equations for the solubility boundary, vapor pressure, surfacetension, dynamic viscosity, thermal conductivity, specific thermal capacity, density,and differential enthalpy of dilution. There is a pore filling threshold where the salt solution leaks from the pores andforms a film on the surface of the silica particles. Tanashev et al. (2013) measuredthe change in the thermal conductivity of salt in silica composites as a function ofadsorbed water and observed a steep rise when the salt solution leaked from thepores, connecting the silica gel particles and enhancing heat transfer. They foundthat for CaCl2 and the other salts tested a steep increase of thermal conductivityoccurred when the calculated volume fraction of the pores occupied with salt solu-tion was 0.60–0.64. As salt solution leakage is detrimental to the sorption chillierperformance, systems should be designed to avoid this regime of operation (Gor-deeva and Aristov, 2012). This paper presents porosimetry and water sorption measurements of CaCl2 con-fined in silica gels with mean pore diameters ranging from 4 to 16 nm. The resultsare compared to the water uptake properties of commercial sorbent materialAQSOA FAM-Z02.

2. EXPERIMENTAL

Chromatography-grade commercial silica gels with four distinct pore size distribu-tions and 0.2–0.5 mm irregular-shaped grains were acquired from Silicycle, Inc.(Quebec, Canada). Concentrated salt solutions were prepared with oven dried CaCl2(Fisher Scientific). The silica gels, in 10 g to 500 g batches, were wetted withethanol, and then an aqueous CaCl2 solution was added to the silica such thatcomplete deposition of the salt into the mesoporous silica would produce the de-sired product. The mixtures dried for 24 h in a fume hood, and 500 g batcheswere further warmed on a hot plate to accelerate drying. The damp material wasbaked at 200oC until judged dry by consistent successive weight measurements.The mesoporous silica, SiliaFlash types B40, B60, B90, and B150, will be referredto hereafter as S4, S6, S9, and S15, while the silica supported salt composites willbe referred to as CaCl2–S4, CaCl2–S6, CaCl2–S9, and CaCl2–S15. The compositeswere 28 wt.% CaCl2 with the exception of CaCl2–S4 (30 wt.%). Salt-in-silica-gel samples consolidated with organic binder polyvinyl pyrrolidone(PVP) (40,000 MW, Sigma Aldrich) and graphite flake (+100 mesh, <150 µm,Sigma Aldrich) as a thermally conductive additive were prepared in small batches.The components were combined in an aqueous solution that was dried on a hotplate, and then heated to 180oC to cross-link the polymer. The product, CaCl2–S15G,was 30% CaCl2, 30% silica, 15% PVP, and 25% graphite flake by weight. Sili-

Volume 6, Issues 3–4, 2015

Evaluation of CaCl2–Silica Gel Sorbent 231

coaluminophosphate desiccant AQSOA FAM-Z02 was acquired from MitsubishiPlastics. This zeolite, referred to hereafter as Z02, was in the form of 1.8–2 mmdiameter pellets containing 6–17 wt.% silicon dioxide binder. Nitrogen isotherms at 77 K were collected with a volumetric gas adsorption ana-lyzer (ASAP 2020, Micromeritics Instrument Corp., USA) to determine the texturalcharacteristics of the silica gels and composites. The samples were degassed undervacuum at 150oC for one hour, followed by two hours at 200oC prior to the meas-urements. The Brunauer, Emmett, and Teller (BET) model was used to calculatethe specific surface area, SBET (Brunauer et al., 1938). Incremental pore volumeper pore width distributions were calculated from the adsorption branch of the iso-therm using the Barrett, Joyner, and Halenda (BJH) model (Barrett et al., 1951). Water sorption isotherms were collected with a thermogravimetric analyzer(TGA) (IGA-002, Hiden Isochema, UK). The samples were dried under vacuum at200oC for six hours to determine their dry weight. Multiple isotherms were col-lected for each sample without repeating the drying process. That is, each isothermended with a final desorption, outgassing step at the isotherm temperature, and thenext isotherm began after the automated system adjusted the sample temperature.The kinetic data for each pressure step were analyzed by a real time processor,and successive pressure steps were taken when water sorption was 98.5% of thepredicted equilibrium. The water uptake was calculated either as

w (gH2O ⁄ gsorbent) = m ⁄ mdry

or

N (moleH2O ⁄ moleCaCl2) = m ⁄ MH2O

(mdry C) ⁄ MCaCl2 ,

where the molar masses, M, of H2O and CaCl2 are 18.15 g/mol and 110.98 g/mol,respectively, and C is the wt.% of salt in the composite. Short-term durability stud-ies of the CaCl2–S15 composite were conducted with the TGA through 200 pres-sure swing cycles at 35oC, whereby the sample was exposed to alternating 12 mbarH2O and vacuum outgassing conditions.

3. RESULTS

The N2 adsorption/desorption isotherms of the silica gels and CaCl2/silica compos-ites were Type IV with H1 hysteresis loops, typical of mesoporous materials. Theisotherms for the CaCl2–S4 and CaCl2–S15 composites and their pure silica gelsupports are shown in Fig. 1, and the calculated textural characteristics are summa-rized in Table 1. Consistent with previous reports, the results indicate that CaCl2fills the silica gel pores and decreases the specific surface area and pore volume.The high uncertainty in pore size distributions calculated from the adsorption


232 McCague et al.

branch of N2 isotherms by the BJH method has been recently discussed by DeLange et al. (2014). The incremental pore volume per pore width curves in Fig. 1provide a qualitative indication of the change in the pore size distribution ofCaCl2–B6 relative to B6, specifically the decrease in pore volume and change inthe pore size distribution due to the preferential filling of small pores. The quali-tative relative differences in the pore sizes distributions of the different silica gelsalt composites prepared for this study are also shown. CaCl2–S4 and CaCl2–S15,the samples with the greatest difference in pore size distribution, were selected fordetailed comparison of their water sorption characteristics.



FIG. 1: Comparison of the N2 isotherms for: a) S4 and CaCl2–S4, b) S15 and CaCl2–S15;comparison of qualitative incremental pore volume per pore width for: c) S4 and CaCl2–S4and d) CaCl2 supported by different silica gels

TABLE 1: Textural characteristics of samples calculated from N2 adsorption isotherms

No.SBET,m2⋅g–1

Porevolume,cm3⋅g–1

Meanpore

diameter,nm

No.SBET,m2⋅g–1

Porevolume,cm3⋅g–1

Meanpore

diameter,nm

S4 574 � 2 0.58 4 CaCl2–S4 146.7 � 0.5 0.27 7S6 487 � 2 0.75 6 CaCl2–S6 134.5 � 0.4 0.37 9S9 406 � 0.4 0.83 8 CaCl2–S9 148.3 � 0.4 0.58 10

S15 276 � 0.8 1.10 16 CaCl2–S15 136.3 � 0.4 0.60 18

The water adsorption branch of the complex isotherms of CaCl2–S4 and CaCl2–S15collected at temperatures from 25 to 80oC are shown in Fig. 2. The initial transitionfrom CaCl2 to CaCl2⋅2H2O (�10 wt.%) is most clearly shown in the 50oC and80oC curves, occurring near 2 mbar and 11 mbar, respectively. The transition issteeper for CaCl2–S15 compared to CaCl2–S4, as is the next increase in curvature oc-curring at �12 mbar on both of the 50oC curves. This distinctive difference betweenthe adsorption of CaCl2 in S4 versus S15 is seen more completely in the 35oCcurves, as a bulge on the CaCl2–S4 isotherm from 4–14 mbar. The equivalent fea-tures are subtler in the CaCl2–S15 isotherm, but the end of this region is visible as asingle point divot at 10.1 and 12.6 mbar in the 25oC and 35oC curves, respectively.This difference is further illustrated by the plots of N (mole H2O per mole CaCl2)as a function of relative pressure, P ⁄ P0, for CaCl2–S4 and CaCl2–S15 at 25oC asshown in Fig. 2c. When plotted as a function of relative pressure, P ⁄ P0, the iso-therms collected at different temperatures were parallel with slightly lower sorptioncapacity with increasing temperature. The offset reflects the strong CaCl2–water in-teractions, however, we have not yet ruled out the possibility of a contribution frominstrumental systematic error, such as an imperfect buoyancy correction.The complexshape of the sorption–desorption isotherm to �0.9P ⁄ P0 at 35oC for CaCl2–S15 isshown in Fig. 2d.


234 McCague et al.

FIG. 2: Water sorption isotherms from 25 to 80oC for: a) CaCl2–S4, b) CaCl2–S15; watersorption isotherms of: c) CaCl2 in S4 and S15 in mole H2O per mole CaCl2 at 25oC;d) CaCl2–S15 to high partial pressure at 35oC

As the salt absorbs water, it transforms into solid CaCl2⋅2H2O, and then deli-quesces. The bulk melting point of CaCl2⋅4H2O is 45oC (Molenda et al., 2013,however the melting point of salt confined in silica is known to be depressed. Fol-lowing the equations of Conde (2004), the density of aqueous CaCl2 solutions canbe calculated as a function of the mole fraction of the salt relative to the water. InFig. 3, the weight of water adsorbed as a function of pressure and the weight ofsalt in the composite are used to estimate the volume of solution in the compositeat various vapor pressures. However, the capillary forces between the solution andporous matrix are known to affect the solution density. On each curve, the vaporpressure at which the upper closure of the hysteresis loop occurs in the watersorption isotherms is marked. At this point, the pores are filled with solution, andthe change of curvature of the isotherm above this point reflects the transition tosorption into the salt solution that has leaked into the interparticle space betweenthe silica gel grains (Fig. 2d). The upper closure point of the hysteresis loops forCaCl2–S4 and CaCl2–S15 occurred at 16 mbar and 40 mbar H2O, respectively,when the calculated volume fraction filled was only 0.8 and 0.74. The upper dashed lines in the plots in Fig. 3 show the pore volume of silica gelhost matrices, S4 and S15, and the lower dashed lines indicate the measured frac-tion of that pore volume filled by CaCl2 in the salt–silica composites. Both thesevalues were determined by nitrogen porosimetry. It is shown that the volume filledby salt far exceeds the volume calculated from the weight of CaCl2 and its bulkdensity, 2.15 g⋅cm–3. The measured volume filled by dry CaCl2 suggests that thedensity of the deposited CaCl2 is low, even comparable to the density of concen-trated aqueous CaCl2 solutions. Alternatively, CaCl2 could be blocking but not fill-ing some of the pores, resulting in the lower pore volume of the composite. The water sorption, pressure, and temperature for 12 mbar pressure swing testsof CaCl2–S15G and Z02 are shown in Fig. 4. During tests, a water bath at 35oC



FIG. 3: Comparison of measured pore volume for: a) S4 and CaCl2–S4, b) S15 andCaCl2–S15 from N2 isotherms with H2O isotherm "pores filled" point (+), and calculatedvolume of CaCl2 (�), CaCl2–2H2O (°), and CaCl2 solution (∆) calculated from the massof sorbed water as a function of pressure and associated CaCl2 molar fraction

surrounds the reactor cylinder and the temperature readings are from a thermome-ter directly adjacent to the sample. From these curves, near equivalent specificcooling powers for cycles to 80% of the sorption capacity, ∆w0.8, were calculatedas W0.8 = ∆w0.8 ∆Hvap ⁄ τ0.8, where τ0.8 is the uptake time and ∆Hvap is the heat ofvaporization of water, 2478 kJ/kg. Under these test conditions, the specific coolingpower of CaCl2–S15G, containing 25 wt.% graphite flakes and 15 wt.% binder,was 0.54 kW/kg compared to 0.60 kW/kg for the benchmark material, Z02, assummarized in Table 2. The sorption performance of CaCl2–S15 was tested for200, 20 min pressure swing cycles at 35oC and found to be consistent with and in-itial and final ∆wcycle of 0.196 and 0.195 g/g. Review of the detailed data indi-cated that both the adsorption step final weight and desorption step final weightincreased slightly during the 200 cycle run, the difference in these increases cre-ated the 0.001 drop in ∆wcycle. The water sorption isobars for CaCl2–S15 and Z02are presented in Fig. 5 along with the equilibrium adsorption times for 0–12 mbarpressure swings at each temperature. It is of note that the uptake rate for Z02 de-clines significantly at �40oC, ahead of the temperature where there is a sharp dropin the uptake capacity. The water sorption isosters for CaCl2–S15 and Z02 are plotted in the ln (PH2O)versus 1 ⁄ T form in Fig. 6 for various amounts of sorbed water, w (g/g). The isosteric heat of water sorption is determined as a function of the slope ofthe linear isosters, where ln (P) = B(w) ⁄ T + C(w) and ∆His(w) = B(w) R and R is


236 McCague et al.

FIG. 4: a) Water sorption as a function of time for a 12 mbar pressure swing for Z02 andCaCl2–S15G (g/g total weight of composite). Test pressure and temperature conditions for;b) CaCl2–S15G; c) Z02

TABLE 2: Uptake, time, and power for pressure swing cycles (0–12 mbar) for 80% equilibriumsorption capacity

Sample T, oC Dry wt.,mg

winitial,g/g

wfinal,g/g

∆w0.8,g/g

τ0.8,min

W0.8,kW/kg

CaCl2–S15G 35 21.231 0.038 0.241 0.163 13.0 0.54Z02 35 19.358 0.036 0.274 0.190 12.5 0.60

the universal gas constant. The increase in ∆His(w) to –63.7 kJ/mol as w ap-proaches 0.09 (N = 2) is due to the formation of solid hydrates with more stronglybound water molecules, and is consistent with reports from Aristov et al. (1996)(62.3 kJ/mol isosteric desorption) for CaCl2 in mesoporous silica gel.

4. CONCLUSIONS

Water sorbent composites of CaCl2 supported by mesoporous silica gels were pre-pared and characterized. The water isotherms of CaCl2 in silica gels with 4 nmand 16 nm average pore sizes were compared. The pore volume filled by solidCaCl2 and concentrated aqueous CaCl2 was measured and calculated for use in fu-ture diffusivity calculations through analysis small pressure step isotherms. Wateruptake capacity of the CaCl2/silica gel sorbent was up to 0.33 g per gram of drysorbent at 12 mbar and 35oC, and sample performance was consistent through 200wetting–drying cycles. The isobars, isosters, and isosteric enthalpy of water sorp-tion of CaCl2–S15 and silicoaluminophosphate desiccant, Z02 were measured andcompared.



FIG. 5: Water sorption isobars at 12 mbar and the equilibrium adsorption time (dt 99%)for 0–12 mbar pressure swings at each temperature for a) CaCl2–S15G (mass % based onthe dry weight of sorbing ingredients, CaCl2, S15, and binder with the graphite flakeweight excluded) and for b) ASQOA FAM-Z02 pellets

FIG. 6: Isosters for a) CaCl2–S15 and b) Z02; c) isosteric water sorption heat for CaCl2–S15and Z02

ACKNOWLEDGMENTS

The authors gratefully acknowledge the financial support of the Natural Sciencesand Engineering Research Council of Canada (NSERC) through Automotive Part-nership Canada Grant No. APCPJ 401826-10. We thank Dr. D. Leznoff and RyanRoberts for their assistance with nitrogen adsorption experiments.

REFERENCES

Aristov, Y., Challenging offers of material science for adsorption heat transformation: A review,Appl. Therm. Eng., vol. 50, pp. 1610–1618, 2013a.

Aristov, Y., Experimental and numerical study of adsorptive chiller dynamics: Loose grains configu-ration, Appl. Therm. Eng., vol. 61, pp. 841–847, 2013b.

Aristov, Y., Tokarev, M., Cacciola, G., and Restuccia, G., Selective water sorbents for multiple ap-plications. 1. CaCl2 confined in mesopores of silca gel: sorption properties, React. Kinet. Catal.Lett., vol. 59, pp. 325–333, 1996.

Barrett, E., Joyner, L., and Halenda, P., The determination of pore volume and area distributions in po-rous substances, computations from nitrogen isotherms, J. Am. Chem. Soc., vol. 73, pp. 373–380,1951.

Brunauer, S., Emmett, P., and Teller, E., Adsorption of gases in multimolecular layers, J. Am. Chem.Soc., pp. 309–319, 1938.

Conde, M., Properties of aqueous solutions of lithium and calcium chlorides: formulations for use inair conditioning equipment design, Int. J. Therm. Sci., vol. 43, pp. 367–382, 2004.

Cullen, J. and Allwood, J., The efficient use of energy: Tracing the global flow of energy from fuelto service, Energy Policy, vol. 38, pp. 75–81, 2010.

De Lange, M., Vlugt, T., Gascon, J., and Kapteijin, F., Adsorptive characterization of porous solids:Error analysis guides the way, Micropor. Mesopor. Mat., vol. 200, pp. 199–215, 2014.

Freni, A., Russo, F., Vasta, S., Tokarev, M., Aristov, Y. I., and Restuccia, G., An advanced solidsorption chiller using SWS-1L, Appl. Therm. Eng., vol. 27, pp. 2200–2204, 2007.

Glaznev, I., Ovoshchnikov, D., and Aristov, Y. I., Effect of residual gas on water adsorption dynam-ics under typical conditions of an adsorption chiller, Heat Transf. Eng., vol. 31, pp. 924–930,2010.

Glaznev, I., Ponomarenko, I., Kirik, S., and Aristov, Y. I., Composites CaCl2/SBA-15 for adsorptivetransformation of low temperature heat: Pore size effect, Int. J. Refrig., vol. 34, pp. 1244–1250,2011.

Gordeeva, L. and Aristov, Y. I., Composites ’salt inside porous matrix’ for adsorption heat transfor-mation: a current state-of-the-art and new trends, Int. J. Low-Carbon Tech., vol. 7, pp. 288–302,2012.

Isaac, M. and van Vuuren, D., Modeling global residential sector energy demand for heating and airconditioning in the context of climate change, Energy Policy, vol. 92, pp. 507–521, 2009.

Molenda, M., Stengler, J., Linder, M., and Worner, A., Reversible hydration behavior of CaCl2 athigh H2O partial pressures for thermochemical energy storage, Thermochimica Acta, vol. 560,pp. 76–81, 2013.

N’Tsoukpoe, K., Restuccia, G., Schmidt, T., and Py, X., The size of sorbents in low pressure sorp-tion or thermochemical energy storage processes, Energy, vol. 77, pp. 983–998, 2014.


238 McCague et al.

Ovoshchnikov, D., Glaznev, I., and Aristov, Y. I., Water sorption by the calcium chloride/silica gelcomposite: The accelerating effect of the salt solution present in the pores, Kinet. Catal., vol. 52,pp. 620–628, 2011.

Perez-Lombard, L., Ortiz, J., and Pout, C., A review on buildings energy consumption information,Energ Buildings, vol. 40, pp. 394–398, 2008.

Sapienza, A., Santamaria, S., Frazzica, A., Freni, A., and Aristov, Y. I., Dynamic study of adsor-bers by a new gravimetric version of the large temperature jump method, Appl. Energy, vol. 113,pp. 1244–1251, 2014.

Tanashev, Y., Krainov, A., and Aristov, Y. I., Thermal conductivity of composite sorbents "salt inporous matrix" for heat storage and transformation, Appl. Therm. Eng., vol. 61, pp. 401–407,2013.

Wang, R. and Oliveira, R., Adsorption refrigeration — An efficient way to make good use of wasteheat and solar energy, Prog. Energ. Combust. Sci., vol. 32, pp. 424–458, 2006.

Zheng, X., Ge, T., and Wang, R., Recent progress on desiccant materials for solid desiccant cooling,Energy, vol. 74, pp. 280–294, 2014.



Documents

EVALUATION OF CaCl 2–SILICA GEL SORBENT FOR WATER …mbahrami/pdf/2015/Evaluation of CaCl2-silica gel sorb… · In the case of salt in silica gel composites, ... 200oC for six